09 August 2012

Getting at the Behavioral Side of Development

While Esther Duflo is often the name most associated with J-PAL up at MIT, it is her co-founder Sendhil Mullainathan who continues to deliver some of the most interesting lectures. In short, Mullainthan is interested in the behavioral side of economics in policy determinations and evaluations. The determination side is the focal point of the talk that he presented recently at the Center for Global Development.

"The diagnoses we have for a problem implicitly or explicitly leads to the solutions we design. And I think the most important aspect of behavioral economics is not clever little designs, it's changing the way that we diagnose what the problem might be," he argues. In doing so, he sets up an interesting conversation going beyond the talk itself.

This was most apparent when David Roodman asked Mullainthan about microfinance in terms of understanding self-control and how to strike a balance of good behavior all around. Rather than discuss what the studies have determined, Mullainthan hones in on the interventions themselves.

Reflecting on his personal work in microfinance, Mullinthan expressed, "I had a bias towards design, and I didn't have enough of a bias towards having an appropriate diagnosis of the underlying problem." He continued to say that the biggest problem of microfinance is that the present model does not appropriately build on the finances of the poor.

What exists, he argues, is a predetermined idea of how microfinance should look like with a wrinkle of behavior change aimed at maximizing positive outcomes. He compares the finances of the poor to juggling where each immediate issue is addressed and tossed into the air only to come back down some time later creating what he calls a 'patchwork' of finance.

That tells him that there is a messy portfolio, rather than a single product. Pay-day loans, to him, address a problem right away but do nothing more than toss another ball into the air.

The video is well worth the 86 minutes. If you are short on time skip ahead to the hour mark for the answer to Roodman's question which gets at the heart of Mullainthan's talk. It leaves me wondering how to realize such a shift in thinking towards understanding the underlying problems. The area becomes far more complex and interesting. Ultimately, how will such a shift in thinking, if possible, change the design of interventions like microfinance?